From e330efca04e2d480b346366c19c4b944f8cc9820 Mon Sep 17 00:00:00 2001 From: Alex Yatskov Date: Sun, 20 Mar 2016 11:05:44 -0700 Subject: Adding source dictionary files --- util/compile.py | 224 ++++++++++++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 224 insertions(+) create mode 100755 util/compile.py (limited to 'util') diff --git a/util/compile.py b/util/compile.py new file mode 100755 index 00000000..7b793ca6 --- /dev/null +++ b/util/compile.py @@ -0,0 +1,224 @@ +#!/usr/bin/env python +# -*- coding: utf-8 -*- + +# Copyright (C) 2013 Alex Yatskov +# +# This program is free software: you can redistribute it and/or modify +# it under the terms of the GNU General Public License as published by +# the Free Software Foundation, either version 3 of the License, or +# (at your option) any later version. +# +# This program is distributed in the hope that it will be useful, +# but WITHOUT ANY WARRANTY; without even the implied warranty of +# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the +# GNU General Public License for more details. +# +# You should have received a copy of the GNU General Public License +# along with this program. If not, see . + + +import codecs +import optparse +import os +import re +import sqlite3 +import sys + + +PARSED_TAGS = { + 'P', # common word + 'adj', # former adjective classification (being removed) + 'adj-f', # noun or verb acting prenominally (other than the above) + 'adj-i', # adjective (keiyoushi) + 'adj-na', # adjectival nouns or quasi-adjectives (keiyodoshi) + 'adj-no', # nouns which may take the genitive case particle `no' + 'adj-pn', # pre-noun adjectival (rentaishi) + 'adj-t', # `taru' adjective + 'adv', # adverb (fukushi) + 'adv-n', # adverbial noun + 'adv-to', # adverb taking the `to' particle + 'aux', # auxiliary + 'aux-adj', # auxiliary adjective + 'aux-v', # auxiliary verb + 'c', # company name + 'conj', # conjunction + 'ctr', # counter + 'exp', # Expressions (phrases, clauses, etc.) + 'f', # female given name + 'g', # given name, as-yet not classified by sex + 'h', # full (usually family plus given) name of a particular person + 'int', # interjection (kandoushi) + 'iv', # irregular verb + 'm', # male given name + 'n', # noun (common) (futsuumeishi) + 'n-adv', # adverbial noun (fukushitekimeishi) + 'n-pref', # noun, used as a prefix + 'n-suf', # noun, used as a suffix + 'n-t', # noun (temporal) (jisoumeishi) + 'num', # numeric + 'p', # place-name + 'pn', # pronoun + 'pr', # product name + 'pref' , # prefix + 'prt', # particle + 's', # surname + 'st', # stations + 'suf', # suffix + 'u', # person name, either given or surname, as-yet unclassified + 'v1', # Ichidan verb + 'v2a-s', # Nidan verb with 'u' ending (archaic) + 'v4h', # Yodan verb with `hu/fu' ending (archaic) + 'v4r', # Yodan verb with `ru' ending (archaic) + 'v5', # Godan verb (not completely classified) + 'v5aru', # Godan verb - -aru special class + 'v5b', # Godan verb with `bu' ending + 'v5g', # Godan verb with `gu' ending + 'v5k', # Godan verb with `ku' ending + 'v5k-s', # Godan verb - iku/yuku special class + 'v5m', # Godan verb with `mu' ending + 'v5n', # Godan verb with `nu' ending + 'v5r', # Godan verb with `ru' ending + 'v5r-i', # Godan verb with `ru' ending (irregular verb) + 'v5s', # Godan verb with `su' ending + 'v5t', # Godan verb with `tsu' ending + 'v5u', # Godan verb with `u' ending + 'v5u-s', # Godan verb with `u' ending (special class) + 'v5uru', # Godan verb - uru old class verb (old form of Eru) + 'v5z', # Godan verb with `zu' ending + 'vi', # intransitive verb + 'vk', # kuru verb - special class + 'vn', # irregular nu verb + 'vs', # noun or participle which takes the aux. verb suru + 'vs-c', # su verb - precursor to the modern suru + 'vs-i', # suru verb - irregular + 'vs-s', # suru verb - special class + 'vt', # transitive verb + 'vz', # Ichidan verb - zuru verb - (alternative form of -jiru verbs) +} + + +def isHiragana(c): + return 0x3040 <= ord(c) < 0x30a0 + + +def isKatakana(c): + return 0x30a0 <= ord(c) < 0x3100 + + +def loadDefinitions(path): + print 'Parsing "{0}"...'.format(path) + with codecs.open(path, encoding='euc-jp') as fp: + return filter(lambda x: x and x[0] != '#', fp.read().splitlines()) + + +def parseKanjiDic(path): + results = list() + + for line in loadDefinitions(path): + segments = line.split() + character = segments[0] + kunyomi = ', '.join(filter(lambda x: filter(isHiragana, x), segments[1:])) + onyomi = ', '.join(filter(lambda x: filter(isKatakana, x), segments[1:])) + glossary = '; '.join(re.findall('\{([^\}]+)\}', line)) + results.append((character, kunyomi, onyomi, glossary)) + + return results + + +def writeKanjiDic(cursor, values): + cursor.execute('DROP TABLE IF EXISTS Kanji') + cursor.execute('CREATE TABLE Kanji(character TEXT, kunyomi TEXT, onyomi TEXT, glossary TEXT)') + cursor.executemany('INSERT INTO Kanji VALUES(?, ?, ?, ?)', values) + + +def parseKradFile(path): + results = list() + + for line in loadDefinitions(path): + segments = line.split(' ') + character = segments[0] + radicals = ' '.join(segments[2:]) + results.append((character, radicals)) + + return results + + +def writeKradFile(cursor, values): + cursor.execute('DROP TABLE IF EXISTS Radicals') + cursor.execute('CREATE TABLE Radicals(character TEXT, radicals TEXT)') + cursor.executemany('INSERT INTO Radicals VALUES(?, ?)', values) + + +def parseEdict(path): + results = list() + + for line in loadDefinitions(path): + segments = line.split('/') + + expression = segments[0].split(' ') + term = expression[0] + match = re.search('\[([^\]]+)\]', expression[1]) + reading = None if match is None else match.group(1) + + glossary = filter(lambda x: len(x) > 0, segments[1:]) + glossary = '; '.join(glossary) + glossary = re.sub('\(\d+\)\s*', str(), glossary) + + tags = list() + for group in re.findall('\(([^\)\]]+)\)', glossary): + tags.extend(group.split(',')) + + tags = set(tags).intersection(PARSED_TAGS) + tags = ' '.join(sorted(tags)) + + results.append((term, reading, glossary, tags)) + + return results + + +def writeEdict(cursor, values): + cursor.execute('DROP TABLE IF EXISTS Terms') + cursor.execute('CREATE TABLE Terms(expression TEXT, reading TEXT, glossary TEXT, tags TEXT)') + cursor.executemany('INSERT INTO Terms VALUES(?, ?, ?, ?)', values) + + +def build(path, kanjidic, kradfile, edict, enamdict): + with sqlite3.connect(path) as db: + if kanjidic is not None: + writeKanjiDic(db, parseKanjiDic(kanjidic)) + + if kradfile is not None: + writeKradFile(db, parseKradFile(kradfile)) + + terms = [] + if edict is not None: + terms += parseEdict(edict) + if enamdict is not None: + terms += parseEdict(enamdict) + if len(terms) > 0: + writeEdict(db, terms) + + +def main(): + parser = optparse.OptionParser() + parser.add_option('--kanjidic', dest='kanjidic') + parser.add_option('--kradfile', dest='kradfile') + parser.add_option('--edict', dest='edict') + parser.add_option('--enamdict', dest='enamdict') + + options, args = parser.parse_args() + + if len(args) == 0: + parser.print_help() + else: + build( + args[0], + options.kanjidic, + options.kradfile, + options.edict, + options.enamdict + ) + + +if __name__ == '__main__': + main() -- cgit v1.2.3